High-performance liquid chromatography (HPLC) is an analytical quantification and detection method in which an eluent is forced through a packed column. The mixture containing the product that has to be detected or quantified is submitted to an eluent flow. Since different products have different elution speeds, the products are separated from each other at the end of the column. After being separated on the column, the product is detected in a flow-through detector. The detector response can be translated into the concentration of the product.
Similarly, in gas chromatography, detectors are used to quantify or detect analytes in a gaseous mixture. Detectors that can be used for these purposes can be the usual electrochemical detectors (ECD). The operation principles of ECD's are well described.
ECD's operate on the basis of potentiometric, amperometric and conductimetric phenomena. The present state of the art will be reviewed for such systems.
Potentiometric detectors are widely used in analytic chemistry, the best known example being the pH- meter. The operating principles are well known. Such detectors are often specific for one ion (ion-selective electrodes). In chromatographic systems (and other systems, based on hydrodynamic measurements) however, potentiometric detection is still poorly developed. For these applications one needs sensors with low specificity, in contrast to batch techniques in which high specificity is required.
The kind of electrodes that are mostly used currently are metal electrodes (copper), so-called liquid-membrane ion-selective electrodes, Ag.sup.+ /AgCl electrodes, and anion-exchange membranes (Potentiometrische detektie van anionen in LC en CE met polymere vloeibaar-membraan elektroden, B. De Backer, Universitaire Instelling Antwerpen, 1995). Such potentiometric sensors are not yet commercially available for liquid chromatographic systems. This is due to the low sensitivity, the slow response and the sometimes irreversible behavior of the sensors under chromatographic conditions.
Liquid membrane electrodes work well under chromatographic conditions as well as in Capillary Zone Electrophoretic (CZE) separations (Grate, M. H. Abraham, Sensors and Actuators B, 3 (1991), 85).
In HPLC one usually uses glassy carbon as the active material in amperometric detectors. Glassy carbon made amperometric detection in HPLC a success in the last 20 years. A wide range of products can be detected with this material. Yet, many chemical compounds can in practice not be oxidized or reduced on the surface of glassy carbon, because of the low reaction rates. Each year about 100 publications report about efforts to modify glassy carbon while improving the kinetics of the oxidation and reduction processes. So far, no significant improvements were made. Another material successfully used in amperometric HPLC detection is polycrystalline platinum, used in the so-called "Pulsed amperometric detection" (PAD) of carbohydrates. Recently, much research is carried out on enzyme-electrodes. The enzymes, immobilized on the electrodes, act as catalysts in the electron transfer from the analyte to the electrode underneath.
Glassy carbon is chemically inert, thereby making developments such as electrode derivatisation or enzyme coupling difficult.
A different application for chemical sensors is in the context of an electronic nose. An electronic nose is an instrument in which an array of sensors, each of which have partial selectivity, is used in combination with a pattern recognition system, and which can be used to recognize simple and/or complex aromas and gas mixtures. Such an array of sensors is described in U.S. Pat. No. 5,571,401.
The electronic nose can be employed in a wide range of applications, for example in industry, medicine, environmental protection, distribution and transport as well as in forensic investigations. In the food and beverage industry one may apply a chemical nose to monitor the freshness, to control quality of the starting, and the middle, and the final products as well as to monitor fermentation processes.
The advantage is that the quasi on-line (in situ) monitoring of the aroma (odor) may lead to an automated control system. Identification and/or quality control of starting products are potential applications in the chemical and pharmaceutical industry, because the "nose" offers a fast and universal analytical method. In medicine, the nose can be used in breath analyses. The patient's breath allows the diagnosis of some diseases, for example the odor of acetone is an indication for diabetes. The odor pattern may be helpful to identify the source of industrial emissions. Yet another potential use is a fast risk analysis in the event of road accidents involving chemicals.
On the spot determinations of fire accelerators may be another application. This list of possible applications is not complete, and other applications will appear. It shows, nevertheless, that the development of the electronic nose is important with many advantageous applications. In several of these applications, a small response time is needed.
The nose must be able--after a learning stage--to distinguish between the aromas of two or more mixtures, each containing a multitude of components. The sensors must operate on a molecular level. Furthermore, a sensor in an electronic nose must have partial selectivities to a wide spectrum of gases, rather than a high sensitivity to only one particular gas. Most existing gas sensors lack the wide spectrum of sensitivities to a variety of gases. Sensors based on metal oxide semi-conductors (MOS) Type 1 sensors, are widely and successfully employed in the sensor industry. Yet in the context of a chemical nose they are inferior compared to other materials. Particularly the number of different MOS materials that can be produced by inorganic synthesis, is very small compared to variability offered in products through organic synthesis.
Other types of sensors considered for application in an electronic nose are all based on the partial, selective permeability of polymers for gasses. On this single physical phenomenon, many technologies can be grafted, such as electrochemical sensors. Also, one knows surface acoustic wave and quartz crystal microbalance sensors using polymer coatings (Dickinson, J. White, J. S. Kauer, D. R. Walt, Nature, 382 (1996), 697) (type 2 sensors). Type 3 sensors exist, based on the fluorescent properties of optical fibers with a polymer coating. Type 4 sensors are conductimetric sensors based on conventional polymers where conductivity is introduced by mixing graphite particles with the polymer.
Conductimetric gas sensors based on conducting polymers form a group of sensors (type 5) that are important in the context of the electronic nose (Persaud, G. Dodd, Nature, 229 (1982), 352). Conducting polymers consist of a long sequence of alternating single and double bonds. This .pi.-conjugated system can be made conductive via oxidation or reduction in a process called doping. The conductivity is influenced by the environment. In other words, conductivity changes upon contact with different vapours. These compounds are well suited for implementation in a chemical nose.
The use of polypyrrole, polyaniline, poly-N-methylpyrrole and poly-5-carboxyindole has been reported. (P. N. Bartlett, S. K. Ling-Chung, Sensors and Actuators, 20, (1989), 287). More type 5 sensors have been made by variation of functional groups on the main skeleton or by a variation of doping materials.
At this moment in time 5 companies offer commercial versions of an electronic nose. They are
(1) Aromascan plc., Electra House, Electra Way, CREWE CW1 1WZ, UK (offering a nose based solely on conducting polymers); PA0 (2) Neotronics, Western House 2, Cambridge Road, Stansted Mountfitchet, Essex CM24 8BZ, UK (offering a nose containing conducting polymer and MOS sensors); PA0 (3) AlphaMOS s.a., 3 Avenue Didier Durat, 31400 Toulouse, France (offering a nose solely based on MOS sensors); PA0 (4) Nordic Sensor Technologies AB, Teknikringen 8, S-583 30 Link'ping, Sweden (offering a nose based on MOS sensors) and PA0 (5) Lennartz Electronic, Bismarckstrasse 136, D-72072 Thbingen, Germany (offering a nose based on MOS and quartz crystal microbalance sensors). PA0 ref. 1: New Scientist, Feb. 18, 1993. PA0 ref. 2: C&EN, Aug. 13, 1996. PA0 ref. 3: Bartlett, J. W. Gardner, "Sensors and Sensory Systems for an Electronic Nose", Kluwer Academic Publisher, Dordrecht, 1992. PA0 ref. 4: Stetter et al., "Sensor systems for an electronic nose.", Kluwer Academic Publischer, Dordrecht, 1992). PA0 ref. 5: Mills, F. Walsh, I. Whyte, Chemical Technology Europe, Jul.-Aug. 26, 1996, PA0 ref. 6: "Selective sample handling and detection in high-performance liquid chromatography", Journal of Chromatography Library, vol. 39A, R. W. Frei and K. Zech, Ed., Elsevier, Amsterdam 1981. PA0 ref. 7: "Principles of ion-selective electrodes and membrane transport", W. E. Morf, Ed., Elsevier, Amsterdam 1981.
Other references to this subject are: